Using split tests to improve results

One of the most effective ways of testing our theories to achieve a high degree of success is in what is called Split Testing. Split Testing does exactly what it sounds like, whatever your idea you test it against an opposing view.

Using web design as an example: perhaps you want to optimise your home page to encourage more people to buy your products - you might have the idea that a change of layout, maybe bringing a big call to action more prominent on the page, will improve the conversion. To test it we create the new layout and only display it to half of the people that visit your website, let it bake for maybe a week at 220 degrees - checking occasionally.

The proof is in the tasting. After the campaign runs for a while, we take a good look at the analytics to see if it has been effective. Often we see only small improvements, and sometimes we need to run a campaign for longer to be sure, but the benefits of continually making small changes and going with what works are immense. Your site evolves naturally, the decisions are based on proof not opinion and the results are always impressive.

Split testing is just one of the ways to make decisions to improve a site, but it is one of the most effective and, maybe we are showing our colours here, one of the most fun to do - maybe that's just us.

If you want to know more leave a comment or (get in touch with us)[/contact-us] - or talk to us on twitter (@gorillacreative)[http://www.twitter.com/GorillaCreative].